Deepfakes have moved from novelty to systemic threat to electoral integrity and personal dignity. Critically evaluate India’s regulatory response under the IT Amendment Rules 2026 and the residual gaps in the criminal-law framework.
Subtopic: Sci-Tech · AI and IT regulation
How to structure your answer
Introduction: MeitY notified the IT (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules 2026 on 10 February 2026 (effective 20 February 2026); over 50 million AI voice-clone calls were estimated during the 2024 Lok Sabha campaign.
Body: 1) Rules framework — definition of "synthetically generated information," visible labels, embedded provenance, takedown clocks of 3 hours (court/government orders) and 2 hours (intimate-image and impersonation deepfakes). 2) Criminal gaps — BNS 2023 covers identity impersonation under Section 319 but no dedicated deepfake offence; lack of victim restitution. 3) Institutional gap — no statutory AI regulator; reliance on Section 79 safe-harbour conditionality.
Way forward: Standalone Digital India Act with risk-tiered AI liability; statutory AI Safety Institute; mandatory C2PA-style content provenance; victim-compensation fund.
Written within the word limit
235 words · target 250 words · 14 min
Introduction:
MeitY notified the IT (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules 2026 on 10 February 2026, effective 20 February 2026; industry trackers estimate over 50 million AI voice-clone calls during the 2024 Lok Sabha campaign. Deepfakes have moved from novelty to systemic threat to electoral integrity, women's dignity, and consumer trust, with the Stanford HAI AI Index 2025 noting India's steepest rise in public AI concern.
The new rules framework:
The Rules define "synthetically generated information," mandate visible labels and embedded provenance (C2PA-style), and impose tight takedown clocks — 3 hours for court or government orders and 2 hours for intimate-image and impersonation deepfakes. Significant social media intermediaries lose Section 79 safe harbour if they default on these timelines, marking India's first binding AI-content regime.
Criminal-law gaps:
The Bharatiya Nyaya Sanhita 2023 covers impersonation under Sections 318-319 and cheating generally, but has no dedicated deepfake offence. Victim restitution, mandatory injunctions, and burden-shifting on platforms remain absent. The Election Commission's 2024 advisory on synthetic media lacks statutory teeth, and the Sadhguru voice-clone scam advisories of 2024 exposed the gap.
Institutional vacuum:
India still lacks a statutory AI regulator. The reliance on Section 79 conditionality leaves enforcement reactive. Delhi High Court's personality-rights orders (Rashmika Mandanna case, November 2023) show judicial creativity but no systemic remedy.
Way forward / Conclusion:
Pass a standalone Digital India Act with risk-tiered AI liability, give the AI Safety Institute statutory backing, mandate C2PA provenance for hosted content, and create a victim-compensation fund under MeitY by 2027.
What an examiner expects to see
- IT Amendment Rules 2026 notified 10 Feb 2026 — first binding AI-content rules
- Takedown clocks: 3 hours for orders, 2 hours for intimate-image and impersonation deepfakes
- ~50 million AI voice-clone calls before 2024 Lok Sabha polls
- Section 79 IT Act 2000 — intermediary safe harbour
- BNS 2023 Sections 318-319 cover cheating and impersonation
- Election Commission advisory on synthetic media 2024
- Stanford HAI 2025 — India saw steepest rise in public AI concern (~14 pp)
Concrete cases, schemes and judgments
- Rashmika Mandanna deepfake case November 2023
- Sadhguru voice-clone scam advisories 2024
- Delhi HC personality-rights orders 2024-25